From: "Martin J. Bligh" <mbligh@mbligh.org>
To: Andi Kleen <ak@suse.de>
Cc: Ingo Molnar <mingo@elte.hu>,
linux-kernel@vger.kernel.org, Andrew Morton <akpm@osdl.org>,
Andy Whitcroft <apw@shadowen.org>
Subject: Re: assert/crash in __rmqueue() when enabling CONFIG_NUMA
Date: Tue, 02 May 2006 09:02:38 -0700 [thread overview]
Message-ID: <4457829E.6090901@mbligh.org> (raw)
In-Reply-To: <200605021745.32907.ak@suse.de>
>>>>Bollocks. It works fine,
>>>
>>>On what kind of box? Some summit system, right?
>>
>>Summit and NUMA-Q, ie everything we originally created it for.
>
> Andi:
>
> That's my point - it usually crashes everywhere else.
Ingo Molnar wrote:
>
> i only booted it on a non-NUMA PC. Most likely the instability is
> caused by some sort of zone mis-sizing. (See more details in this
> same thread.)
Ooooh, on ordinary PCs. that makes more sense.
> The problem is that it's not regression tested and quite complex and
> tends to break often and stay broken.
OK, well the regression testing issue is easily fixed, but whether it's
worth it or not is a different issue. It was originally done for the
distros really, so they could have a single kernel that supported
everything.
> If you don't want to mark it CONFIG_BROKEN then i would suggest a panic
> early when the system isn't SUMMIT (I think NUMAQ does this already)
>
> Something like the appended patch
apw: this was your baby ... what do you want to do with it? Add it
to the automated regression testing, or kill it?
> -Andi
>
> i386: Panic the system early when a NUMA kernel doesn't run on IBM NUMA
>
> It has been broken forever anywhere else and is not too useful
> anyways so best to disable it.
>
> Signed-off-by: Andi Kleen <ak@suse.de>
>
> Index: linux/arch/i386/kernel/srat.c
> ===================================================================
> --- linux.orig/arch/i386/kernel/srat.c
> +++ linux/arch/i386/kernel/srat.c
> @@ -327,6 +327,14 @@ int __init get_memcfg_from_srat(void)
> int tables = 0;
> int i = 0;
>
> + extern int use_cyclone;
> + if (use_cyclone == 0) {
> + /* Make sure user sees something */
> + static const char s[] __initdata = "Not an IBM x440/NUMAQ. Don't use i386 CONFIG_NUMA anywhere else."
> + early_printk(s);
> + panic(s);
> + }
> +
> if (ACPI_FAILURE(acpi_find_root_pointer(ACPI_PHYSICAL_ADDRESSING,
> rsdp_address))) {
> printk("%s: System description tables not found\n",
> Index: linux/arch/i386/Kconfig
> ===================================================================
> --- linux.orig/arch/i386/Kconfig
> +++ linux/arch/i386/Kconfig
> @@ -517,6 +517,9 @@ config NUMA
> depends on SMP && HIGHMEM64G && (X86_NUMAQ || X86_GENERICARCH || (X86_SUMMIT && ACPI))
> default n if X86_PC
> default y if (X86_NUMAQ || X86_SUMMIT)
> + help
> + NUMA support. Note this only works on IBM x440 or IBM NUMAQ.
> + Don't try to use it anywhere else.
>
> comment "NUMA (Summit) requires SMP, 64GB highmem support, ACPI"
> depends on X86_SUMMIT && (!HIGHMEM64G || !ACPI)
next prev parent reply other threads:[~2006-05-02 16:02 UTC|newest]
Thread overview: 68+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20060419112130.GA22648@elte.hu>
2006-04-20 9:18 ` Nick Piggin
2006-04-21 11:20 ` Ingo Molnar
2006-04-21 11:45 ` Ingo Molnar
2006-05-01 12:49 ` Ingo Molnar
2006-05-02 6:48 ` Andi Kleen
2006-05-02 7:06 ` Ingo Molnar
2006-05-02 7:05 ` Andi Kleen
2006-05-02 8:27 ` Ingo Molnar
2006-05-02 14:02 ` Martin J. Bligh
2006-05-02 14:25 ` Nick Piggin
2006-05-04 1:32 ` Bob Picco
2006-05-04 8:37 ` Ingo Molnar
2006-05-04 9:14 ` Ingo Molnar
2006-05-04 9:26 ` Ingo Molnar
2006-05-04 8:37 ` Andy Whitcroft
2006-05-04 15:21 ` Dave Hansen
2006-05-04 15:46 ` Bob Picco
2006-05-04 16:07 ` Dave Hansen
2006-05-04 19:25 ` Ingo Molnar
2006-05-04 19:43 ` Bob Picco
2006-05-04 21:50 ` Andy Whitcroft
2006-05-05 5:17 ` Ingo Molnar
2006-05-05 13:55 ` Bob Picco
2006-05-05 14:33 ` Dave Hansen
2006-05-05 14:50 ` Bob Picco
2006-05-05 14:57 ` Dave Hansen
2006-05-05 15:03 ` Martin J. Bligh
2006-05-05 16:22 ` Bob Picco
2006-05-05 16:18 ` Bob Picco
2006-05-06 8:32 ` Nick Piggin
2006-05-07 13:07 ` Andy Whitcroft
2006-05-07 13:18 ` Nick Piggin
2006-05-09 11:05 ` [PATCH 0/3] Zone boundry alignment fixes Andy Whitcroft
2006-05-09 11:05 ` [PATCH 1/3] zone init check and report unaligned zone boundries Andy Whitcroft
2006-05-09 11:28 ` Nick Piggin
2006-05-09 11:05 ` [PATCH 2/3] x86 align highmem zone boundries with NUMA Andy Whitcroft
2006-05-09 11:05 ` [PATCH 3/3] zone allow unaligned zone boundries Andy Whitcroft
2006-05-11 7:59 ` [PATCH 0/3] Zone boundry alignment fixes Andrew Morton
2006-05-12 14:19 ` Ingo Molnar
2006-05-13 1:39 ` Nick Piggin
2006-05-18 14:20 ` [PATCH 0/2] Zone boundary alignment fixes cleanups Andy Whitcroft
2006-05-18 14:21 ` [PATCH 1/2] zone init check and report unaligned zone boundaries fix Andy Whitcroft
2006-05-18 14:21 ` [PATCH 2/2] zone allow unaligned zone boundaries spelling fix Andy Whitcroft
2006-05-18 14:49 ` Andy Whitcroft
2006-05-18 15:54 ` [PATCH 0/2] Zone boundary alignment fixes, cleanups v2 Andy Whitcroft
2006-05-18 15:55 ` [PATCH 1/2] zone init check and report unaligned zone boundaries fix Andy Whitcroft
2006-05-18 15:55 ` [PATCH 2/2] zone allow unaligned zone boundaries spelling fix Andy Whitcroft
2006-05-02 15:03 ` assert/crash in __rmqueue() when enabling CONFIG_NUMA Andi Kleen
2006-05-02 15:17 ` Martin J. Bligh
2006-05-02 15:45 ` Andi Kleen
2006-05-02 16:02 ` Martin J. Bligh [this message]
2006-05-02 16:05 ` Andi Kleen
2006-05-02 19:47 ` Ingo Molnar
2006-05-02 19:48 ` Ingo Molnar
2006-05-02 19:44 ` Andi Kleen
2006-05-02 19:56 ` Martin Bligh
2006-05-02 20:00 ` Andi Kleen
2006-05-02 20:13 ` Ingo Molnar
2006-05-02 20:12 ` Andi Kleen
2006-05-02 15:52 ` Ingo Molnar
2006-05-02 19:55 ` [RFC, PATCH] cond_resched() added to close_files() Eric Dumazet
2006-05-03 7:01 ` Ingo Molnar
2006-05-12 9:44 ` Andrew Morton
2006-05-12 10:20 ` Ingo Molnar
2006-05-12 12:24 ` Eric Dumazet
2006-05-14 0:09 ` Lee Revell
2006-04-21 11:51 ` assert/crash in __rmqueue() when enabling CONFIG_NUMA Nick Piggin
2006-04-19 11:23 Ingo Molnar
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4457829E.6090901@mbligh.org \
--to=mbligh@mbligh.org \
--cc=ak@suse.de \
--cc=akpm@osdl.org \
--cc=apw@shadowen.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
Powered by JetHome